Kentucky - Vol. #2v2

My analysis yesterday was wrong, since this was an OR not an RR comparison, but Iam pretty sure my were not.

Ok, in the Kentucky study, the rates of Fully-Vaccinated in the Cases and Controls are way too close.

I will try to show what I mean through a simple model.

For this model, I am excluding partially vaccinated pts.

Using the Kentucky study's stats, the Controls have a Fully-Vaccinated (FV) rate = 34%. I will use that and I will use a Never-Vaccinated (NV) rate = 66%.

Assume 1,000 Controls at the beginning of the risk period.

At the start of the risk period, there are 660 NV and 340 FV pts.

Assume any infection rate you want, that rate washes out.

In the first table below, I used an infection rate for NV pts = 10%, meaning during this period, 10% of NV pts would become infected and then, in the second table, I used an infection rate for NV pts = 1% to show the rate doesn't matter.

Also, assume some level of vaccine protection. I used a protection rate = 90%. This rate DOES matter.

Ok, so at the end of the risk period, 10% of 660 NV developed infection or 66 pts.

And the infection rate of FV pts is 1/10 of 10% = 1%, so 3.4 FV pts developed infection.

The FV percentage falls from 34% of the uninfected group at the start to 4.9% in the infected group. This change is not affected by the absolute NV infection rate.

The ratio of FV rates is 34%/4.9% = 8.7.

In the study, the FV rate ratio of Controls:Cases was 34%/21%= 1.6.

Ok, it gets worse.

As shown in Tables 1 & 2, the absolute infection rate has no effect on the change in FV percentage.

However, the relative rate of infection very much does and is based upon the efficacy of the vaccines.

In the Kentucky study, the Cases, the Infected Group, had a FV rate = 21%.

For the FV rate to fall from 34% in the Uninfected Group to 21% in the Infected Group, the relative rate of infection of the FV pts to the NV pts has to equal ~50%.

Said another way, the vaccine's efficacy has to = 50% and nothing close to 90%.

So, something's wrong down in Kentucky.

Either the vaccine is only 50% effective or something else.

Again, maybe I am missing something

But I don't see how two, very closely related groups, differing by Covid infection during May and June 2021 have such similar fully-vaccinated rates.

SMS

P.S. – I have not heard back from the communicating author. Perhaps, her title is ironic.